Fast beam migration — A step toward interactive imaging

Gao, Fuchun; Zhang, Po; Wang, Bin; Dirks, Volker

doi:10.1190/1.2370032

Cited by 30 publications

(9 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the case of equation 13, despite the fact that only the reflection angle information is required, such reflection angles depend on dip, and thus, reflector dip information or at least the reflection slope is also needed. This suggests implementations synonymous to fast-beam methods (Gao et al, 2006), in which the wavefield is locally slant stacked along the midpoint axis and only the most energetic of plane waves are extrapolated. Thus, like fast-beams methods, these equations could allow for crude and fast extrapolation of wavefields by using data slope information.…”

Section: Discussionmentioning

confidence: 99%

Prestack wavefield approximations

Alkhalifah

2013

GEOPHYSICS

View full text Add to dashboard Cite

The double-square-root (DSR) relation offers a platform to perform prestack imaging using an extended single wavefield that honors the geometrical configuration between sources, receivers, and the image point, or in other words, prestack wavefields. Extrapolating such wavefields, nevertheless, suffers from limitations. Chief among them is the singularity associated with horizontally propagating waves. I have devised highly accurate approximations free of such singularities which are highly accurate. Specifically, I use Padé expansions with denominators given by a power series that is an order lower than that of the numerator, and thus, introduce a free variable to balance the series order and normalize the singularity. For the higher-order Padé approximation, the errors are negligible. Additional simplifications, like recasting the DSR formula as a function of scattering angle, allow for a singularity free form that is useful for constant-angle-gather imaging. A dynamic form of this DSR formula can be supported by kinematic evaluations of the scattering angle to provide efficient prestack wavefield construction. Applying a similar approximation to the dip angle yields an efficient 1D wave equation with the scattering and dip angles extracted from, for example, DSR ray tracing. Application to the complex Marmousi data set demonstrates that these approximations, although they may provide less than optimal results, allow for efficient and flexible implementations.

show abstract

Section: Discussionmentioning

confidence: 99%

Prestack wavefield approximations

Alkhalifah

2013

GEOPHYSICS

View full text Add to dashboard Cite

show abstract

“…Current computer hardware allows us to create multiple realizations of 3D seismic imaging results with different models within a couple of days, even with standard Kirchhoff implementations. Faster approaches like beam-type methods ͑Hill, 1990, Gao et al, 2006͒ show promise in allowing true real-time model-image interaction. Fast-turnaround PSDM allows the interpreter, seismic processor, and other researchers to explore possible solutions to the imaging problem and gain a deeper understanding of uncertainties related to the basin under investigation.…”

Section: Discussionmentioning

confidence: 99%

Geologically constrained seismic imaging — Workflow

et al. 2008

View full text Add to dashboard Cite

We present the geologically constrained workflow for velocity-model building as a case study from offshore Brazil. The workflow involves basin reconstruction, gravity modeling, and seismic interpretation in addition to standard prestack depth migration ͑PSDM͒ model-building tools. Building a salt model based on seismic evidence can be highly nonunique. In a geologically constrained seismic-processing workflow, the main aim is to use geologic understanding with geophysical models and datasets to improve an input velocity realization for the PSDM loop, thereby improving image quality. All of these methods are inherently uncertain, and a final model is based on a range of subjective choices. Thus a final result that agrees with all sciences still can be completely wrong. However, an understanding of these choices enables a unique way of testing and constraining the number of antimodels: velocity models that fit the observations but are different from the final result. This can reduce time spent and uncertainty in geologic evaluation.

show abstract

“…On the other hand, only picking a few dominant ray parameters for imaging can be benefit for implementation efficiency. The fast beam migration (Gao et al, 2006) is an application of the same idea, but it has no consideration of the characteristic seismic phase separation. In other words, picking some dominant ray parameters in the traditional local τ − p domain is one of characteristic expressions of the original seismic data (Stoffa et al, 1981), but it has no firm theoretical basis and the corresponding result is not sparse enough.…”

Section: Introductionmentioning

confidence: 98%